Warning: Creating default object from empty value in /home/johnnz/public_html/wp-content/themes/simplicity/functions/admin-hooks.php on line 160

Data Quality: Fatally Flawed

Data Quality is a Fatally Flawed Approach

The trouble with Data Quality is that it is all about data – and this is a fatal flaw!!

Managing data in any enterprise as a standalone element with an imagined intrinsic value is a fundamentally flawed approach that brings no sustainable benefits of any kind.

Double Whammy

The current Data Quality approach is flawed in two major ways.

Firstly, managing the quality of data alone is no guarantee of the quality of the information required by the enterprise.  It is information, not data, that is the lifeblood of an enterprise and information has two elements, data and structure.

Managing the quality of  bricks will not turn them into a building.

All of these bricks may conform to the defined quality standards but they are not a building.

Bricks are not a building. Data is not information.


This building is a structure that is fit for the function for which it was designed (its purpose), making it a quality building, yet all of its bricks may not be perfect.

Data = Brick | Information = Building

The second major way in which Data Quality is flawed is that the data it deals with has no purpose.  Without purpose it is not possible to define or achieve quality. Fitness for purpose is the cornerstone of the definition of quality in every sphere of activity around the world – apart from in the Data Quality world it would seem!  Is this true? No, it’s a myth!

A DQ World Without Purpose!

It’s a myth created by those Data Quality practitioners who are working in a world without purpose and, rather than admit that their approach is flawed, have decided to claim that their world is so unique that purpose is not relevant!! So, so wrong!  Yet so simple to rectify!

There is one purpose for all data in every enterprise and that is to support the execution of the Business Functions of that Enterprise.  When you know what the Business Functions are, then you can define all of the data required by the enterprise and the structure of that data.  Data with a defined structure is called Information!

Without knowing the Business Functions you cannot know what information is required by the enterprise and so can never effectively manage the data that ought to form part of that information.

Triple & Quadruple Whammy

The current DQ approach also gives rise to two further myths, 1) that Real World Alignment replaces Fitness for Purpose and 2) the illusion of Data Re-use.  But more about those in another post.

So what is the Answer?

Information Quality! A practice dedicated to providing the enterprise with all of the information required by the Business Functions of that enterprise to the standards defined by those Business Functions.

It makes sense.  After all, Information Quality is the true IQ of every enterprise!

Share the Love

If you think that this post might be of interest to a colleague or friend please share it with them.


8 Responses to “Data Quality: Fatally Flawed”

  1. Perry Rogers October 17, 2012 7:04 pm #

    Hi John

    Over the years as I’ve grappled with my various roles in the Data Governance and Quality space I’ve learned that DQ, Information Design-Delivery, and Process improvement are all closely joined at the hip. And my customers don’t know the difference, and they don’t really care to know. They just look at me to fix it, and will hold me accountable for it.

    All too often in Data Governance roles I’ve found myself having to drive true root cause mitigation of data/information fitness issues by stepping beyond my role and actually venturing into spaces of accountability that aren’t actually mine, because that is where the RC lives. Typically these are Quality Control in-process issues, and accountability issues.

    I believe there is definitely a place for DQ and DG but in the absence of a business culture that accepts accountability for In-process Quality Control of data, and ownership of processes, the data steward is on a hiding-to-nothing to achieve their objectives and may end-up all too often fixing problems that aren’t really theirs, yet manifest the issues that business are having with their data.

    Just my two cents :)


    • John Owens October 20, 2012 10:31 pm #

      Hi Perry

      Thanks for your comment. The root cause will nearly always lie upstream from the area in which we are working and, as such, will often lie outside our strict remit!

      This is a dilemma that I have often faced. It is hard to know what to do. Just do what I am contracted to do, take the money and don’t rock the boat, even though this will bring no real benefits to my clients?

      Or do I poke my nose in where it does not belong?

      What I have come to realise is that my nose actually it does belong there. Because, if I do not address upstream root cause issues, then I am misleading my clients in letting them believe that my efforts will bring them any real benefits.

      Sometimes I have had to walk away from an assignment and tell them that I cannot help them without broadening my remit – a painful financial decision! However, in the long run I have benefitted from this as I have learned ways in which to help clients realise that they themselves must look upstream to see and remove the root cause of their quality problems. This is the essence of quality assurance, to prevent defects arising in the first place.


  2. Rubina September 21, 2012 12:02 pm #

    I like the critical approach, but I am not sure that Information Quality is the answer. Could you please elaborate? Information Quality is alo context dependent, is it not?

    • John Owens September 21, 2012 8:52 pm #

      Hi Rubina

      Thank you for your feedback.

      My view is simple. The key activities of any enterprise are its Business Functions, they are in fact its raison d’etre.

      In order to effectively carry out the Business Functions, information with defined data elements and a define structure is required. Information meeting these criteria can be defined as ‘Quality’ Information as it conforms to requirements, i.e to support the Business Functions of the enterprise.

      This view, though simple, is very powerful as it enables Information Quality to be clearly defined, understood and, because it is now understood, achieved.


  3. Richard Ordowich September 20, 2012 9:46 am #

    Not sure what the argument is? Product quality is a sum total of all business policies, processes, raw materials, customer expectations, capabilities data, information etc.

    Isolating one aspect of quality be it data, information or process quality is limiting. Unless the organization adopts a quality culture, isolated data quality or information quality efforts will be limited.

    Quality has to be adopted systemically. Arguing whether its data, information or process quality is fruitless. It’s all of the above!

    • John Owens September 20, 2012 10:32 am #

      Hi Richard,

      The argument is that all of the effort of the Data Quality community is going into trying to control the quality of data and just data. This in itself is a flawed approach. To compound this, a large part of this community has abandoned the fundamentals of quality by a) being unwilling to or b) being unable to define the fitness for purpose of what they are trying to manage.

      You cannot abandon the fundamentals of Quality and expect to achieve it, no matter how much time, effort, money or rhetoric you throw at it.

  4. James September 20, 2012 5:59 am #

    Isn’t data quality loosely defined as “fit for purpose” and data leads to information leads to knowledge leads to wisdom?

    Your analogy of the house is a good one, not only do the bricks need to be fit for purpose but also the mortar, the foundations, the design (plan) of the dwelling and so on. Bricks ideal for a house (face bricks look nice) might crumble to dust in a kiln due to enormous heat and pressure. Thus a good quality brick depends on it’s purpose.

    Check our this blog and tell me what you think………


    • John Owens September 20, 2012 6:26 am #

      Hi James

      Many Data Quality practitioners pooh-pooh the term ‘fit for purpose’, claiming it does not apply to data!! They really are missing the point and fighting a losing battle.

      However, DQ is now a multi-million dollar industry which, if information quality was properly implemented, would cease to exist. So…

Leave a Reply